OView-AI Supporter for Classifying Pneumonia, Pneumothorax, Tuberculosis, Lung Cancer Chest X-ray Images Using Multi-Stage Superpixels Classification

Abstract

The deep learning approach has recently attracted much attention for its outstanding performance to assist in clinical diagnostic tasks, notably in computer-aided solutions. Computer-aided solutions are being developed using chest radiography to identify lung diseases. A chest X-ray image is one of the most often utilized diagnostic imaging modalities in computer-aided solutions since it produces non-invasive standard-of-care data. However, the accurate identification of a specific illness in chest X-ray images still poses a challenge due to their high inter-class similarities and low intra-class variant abnormalities, especially given the complex nature of radiographs and the complex anatomy of the chest. In this paper, we proposed a deep-learning-based solution to classify four lung diseases (pneumonia, pneumothorax, tuberculosis, and lung cancer) and healthy lungs using chest X-ray images. In order to achieve a high performance, the EfficientNet B7 model with the pre-trained weights of ImageNet trained by Noisy Student was used as a backbone model, followed by our proposed fine-tuned layers and hyperparameters. Our study achieved an average test accuracy of 97.42%, sensitivity of 95.93%, and specificity of 99.05%. Additionally, our findings were utilized as diagnostic supporting software in OView-AI system (computer-aided application). We conducted 910 clinical trials and achieved an AUC confidence interval (95% CI) of the diagnostic results in the OView-AI system of 97.01%, sensitivity of 95.68%, and specificity of 99.34%.

Keywords: EfficientNet; deep learning; lung cancer; pneumonia; pneumothorax; tuberculosis.

Figure 1

Overall study flow of dataset collecting and labeling for the multi-classification for lung diseases.

Figure 2

Composition of dataset and test dataset.

Figure 3

Lung diseases dataset: (a) normal; (b) pneumonia; (c) pneumothorax; (d) tuberculosis; (e) lung cancer.

Figure 4

Overall pipeline of multi-classification for lung diseases using transfer learning technique.

Figure 5

EfficientNet B7 architecture as our backbone model.

Figure 6

Our proposed fine-tuned model.

Figure 7

Training and validation performance: (a) accuracy; (b) sensitivity; (c) specificity.

Figure 8

Results of inferencing on 918 images: (a) receiver operating characteristic (ROC) curve; (b) confusion matrix.

Figure 9

Results of inferencing on 910 images: (a) receiver operating characteristic (ROC) curve; (b) confusion matrix.

Figure 9

Results of inferencing on 910 images: (a) receiver operating characteristic (ROC) curve; (b) confusion matrix.